WO2009064562A2

WO2009064562A2 - Spatial separation in a relay communication system

Info

Publication number: WO2009064562A2
Application number: PCT/US2008/079700
Authority: WO
Inventors: Sebastien Simoens
Original assignee: Motorola, Inc.
Priority date: 2007-11-09
Filing date: 2008-10-13
Publication date: 2009-05-22
Also published as: WO2009064562A3; GB2454657B; GB2454657A; GB0722042D0

Abstract

An apparatus (201) comprises a quantizer (211) receiving first signals received by a first antenna array (209) from a plurality of remote stations (203). A first relay station (205) receives second signals from the remote stations (203) at a second antenna array (215) and performs a data reduction processing before forwarding the data to the apparatus (201). A spatial processor (401) performs a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals to generate received data streams for a set of remote stations (203) of the plurality of remote stations (203). The processing provides a spatial separation for the received data streams from the set of remote stations (203) and is performed in response to a data rate parameter indicative of the data rate reduction processing performed at the first relay (205). The data rate reduction may include quantization and compression. Improved performance can be achieved by the combination of relaying and spatial separation.

Description

CML04885M SPATIAL SEPARATION IN A RELAY COMMUNICATION SYSTEM

Field of the invention

The invention relates to a spatial separation in a relay communication system and in particular, but not exclusively, to Spatial Division Multiple Access (SDMA) in a cellular communication system employing relay techniques.

Background of the Invention

In some wireless radio communication systems, it has been proposed to use intermediate relay stations to improve or facilitate communication between two communication units. Specifically, communication from remote stations to a centralised base station or access point may be supported by intermediate relay stations in various communication systems such as e.g. IEEE 802.16j or IEEE 802.11s.

Moreover, it has also been proposed that in a coordinated cellular network, a centralised base station may be supported by another base station. Thus, such a supporting base station is a relay for the destination base station. Thus, a relay can include a supporting relay station or base station, connected to a centralised base station via either a wireless link or a wired link, and which facilitates the communication from remote stations to the centralised base station. CML04885M

In a simple relay system, an intermediate relay may simply relay radio signals in a simple linear relay path where any given communication unit in the path only processes the signal from the previous relay in the relay path. Such a system is known as a non-cooperative relaying system and is typically useful when the range of the destination communication needs to be extended.

A more complex system is known as cooperative relaying wherein each communication unit may process the signals from several predecessors such that these cooperate in order to maximize the information rate transmitted from the source to the destination device. FIG. 1 illustrates a simple example of cooperative relaying involving a single relay R. In the example, the relay R receives the signal from the source device S and forwards this to the destination device D. In addition, the destination device D also directly receives the signal from the source device S. The destination device D then performs a joint reception taking into account both the direct and the relayed signals. Thus, the relay R cooperates with the destination device D and the source device S to improve the communication.

Various cooperative relaying techniques have been proposed including the Decode-and-Forward (DF) technique and the Compress-and-Forward (CF) technique. In general, cooperative DF tends to perform better when the relay is close to the source device and cooperative CF tends to perform better when the relay is close to the destination device . ° CML04885M

The DF technique relies on the use of regenerative relays wherein the received data is fully decoded before the data is transmitted to the destination.

In contrast, for the cooperative CF technique the relay does not decode the data received from the source but instead quantizes the received signal samples and compresses them using a lossy compression technique. The compressed samples are then channel-encoded and transmitted from the relay to the destination. The destination then performs channel-decoding and decompression of the data to regenerate the quantized samples. Accordingly, the destination has two observations or signal sets available to perform the final data decoding. Basically, the destination has both the signal received from the source by the relay as well as the signal received directly from the source. The signal from the relay may be distorted due to the quantization and lossy compression processes. However, these processes allow the amount of data transmitted from the relay to the destination to be reduced to practical levels. The decoding of the data at the destination device taking into account both the direct and relayed signals provide an improved reception thereby allowing an improved utilisation of the available radio resource and an increased capacity of the communication system.

Another technique that has been proposed for improving communication systems is the use of spatial separation between users. Specifically, some radio communication systems use Spatial Division Multiple Access wherein the same the same time-frequency (-code) resource (sometimes referred to as a "resource chunk") of a centralised base CML04885M station or access point can be allocated to more than one user if these are spatially separated.

Specifically, the base station or access point can contain an antenna array comprising a plurality of antenna elements. A beamforming algorithm is then applied to generate directional beams in different directions corresponding to the scheduled users for the time-frequency resource block. By beamforming in the direction of a desired remote station and creating nulls in the direction of other remote stations, an effective separation can be achieved thereby allowing a plurality of users to share the same time-frequency resource block.

Thus, spatial multiplexing separates remote stations in the spatial domain in order to achieve a spatial multiplexing gain. The total information rate that can be received from the scheduled and spatially separated remote stations (known as the sum-rate) can substantially exceed the information rate that can be received from a single remote station allocated the unshared time- frequency resource block. Therefore, a higher spectral efficiency and capacity of the system can be achieved. The maximum number of single antenna remote stations that can be scheduled simultaneously must be lower than the number of antennas at the base station or access point in order for this to generate a separate beam for each remote station. However, the number of simultaneous remote stations is also limited by the spatial channel characteristics for each remote station and by the processing capabilities of the base station or access point. In practice, the throughput that can be achieved depends on the specific receiver processing and the r.

° CML04885M separability of the channels for the remote stations that share a time-frequency (-code) resource block. It has been proposed to combine SDMA with systems using relays. However, the suggested approaches tend to be suboptimal and result in cumbersome and/or impractical implementation and/or degraded performance. For example, in order to provide both SDMA and relaying in the uplink, the proposed systems generally involve performing a spatial separation at the relay in connection with a decoding of the received signals. Thus, each relay performs individual beamforming and decoding of the received signals and then forward the decoded data to the destination. However, the use of such a Decode-and- Forward (DF) approach tends to be suboptimal in many situations, such as when the relay is close to the destination. Furthermore, the spatial separation of remote stations is limited by the number of antennas at the relay.

Hence, an improved system would be advantageous and in particular a system allowing increased flexibility, facilitated implementation, improved spatial separation, an increased ability to spatially separate more users, increased spectral efficiency and/or improved performance would be advantageous .

Summary of the Invention

Accordingly, the Invention seeks to preferably mitigate, alleviate or eliminate one or more of the above mentioned disadvantages singly or in any combination. c

° CML04885M

According to an aspect of the invention there is provided an apparatus comprising: means for receiving first signals received by a first antenna array from a plurality of remote stations; means for receiving, from a first relay station, second signals received by a second antenna array of the first relay station from the plurality of remote stations, the second signals being represented by data dependent on a data rate parameter indicative of a data rate reduction processing performed at the first relay station; means for performing a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals to generate received data streams for a set of remote stations of the plurality of remote stations, the spatial separation joint processing providing a spatial separation for the received data streams; and wherein the spatial separation joint processing is in response to the data rate parameter.

The invention may provide improved performance in many wireless communication systems. Specifically, an improved spectral efficiency and/or increased capacity may be achieved. The invention may e.g. allow an increased number of beams to be generated and used for spatial separation of remote stations. The number of beams may be limited by the total number of antennas in the first and second antenna array rather than by the number of antennas in one array.

The invention may allow an improved system using both

SDMA and relaying. In particular, the invention may allow SDMA to be used with a Compress and Forward relaying approach. CML04885M

According to another aspect of the invention, there is provided a method comprising: receiving first signals received by a first antenna array from a plurality of remote stations; receiving, from a first relay station, second signals received by a second antenna array of the first relay station from the plurality of remote stations, the second signals being represented by data dependent on a data rate parameter indicative of a data rate reduction processing performed at the first relay station; performing a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals to generate received data streams for a set of remote stations of the plurality of remote stations, the spatial separation joint processing providing a spatial separation for the received data streams; and wherein the spatial separation joint processing is in response to the data rate parameter.

These and other aspects, features and advantages of the invention will be apparent from and elucidated with reference to the embodiment (s) described hereinafter.

Brief Description of the Drawings

Embodiments of the invention will be described, by way of example only, with reference to the drawings, in which

FIG. 1 is an illustration of a relay communication system in accordance with the prior art; ° CML04885M

FIG. 2 illustrates an example of a cellular communication system in accordance with some embodiments of the invention;

FIG. 3 illustrates an example of a cellular communication system in accordance with some embodiments of the invention;

FIG. 4 illustrates an example of a data receiver in accordance with some embodiments of the invention;

FIG. 5 illustrates an example of a parameter processor in accordance with some embodiments of the invention; and

FIG. 6 illustrates an example of a method of receiving data from remote stations in accordance with some embodiments of the invention.

Detailed Description of Some Embodiments of the Invention

The following description focuses on embodiments of the invention applicable to a cellular communication system such as an IEEE 802.16j or IEEE 802.11s cellular communication system. However, it will be appreciated that the invention is not limited to this application but may be applied to many other communication systems using relay techniques.

FIG. 2 illustrates elements of a cellular communication system in accordance with some embodiments of the invention . Q

^Ό CML04885M

The cellular communication system comprises a base station 201 (or access point) which supports a plurality of remote stations over the air interface of the communication system. Specifically, the base station can support all remote stations which are within the cell served by the base station 201.

FIG. 2 illustrates a single remote station 203 which for example may be a user equipment, a mobile phone, a remote terminal, a subscriber unit, a 3G User Equipment or any other communication unit or entity capable of communicating over an air interface of the communication system. The source communication unit 203 is served by the base station 201.

It will be appreciated that although FIG. 2 for brevity and clarity illustrates only a single remote station 203 served by the base station 201, the base station 201 will be capable of simultaneously supporting a plurality of communication units over the air interface.

The cellular communication system of FIG. 2 furthermore comprises a relay station 205 which is operable to receive signals from the remote station 203 and forward these to the base station 201. The communication system specifically uses cooperative relaying based on a Compress-and-Forward approach.

In the following, the operation of the base station 201, remote station 203 and relay station 205 for a transmission from the remote station 203 to the base station 201 will be described. Thus, the remote station 203 is the source communication unit for the transmission ^ CML04885M and the base station 201 is the destination communication unit for the transmission.

The remote station 203 comprises an antenna array 207 from which the uplink signal is transmitted in a first time slot (FIG. 2 illustrates an antenna array containing two antenna elements but it will be appreciated that any number of antenna elements may be used in the array including a single antenna) .

The transmitted signal is received by an antenna array 209 coupled to the base station 201. The antenna array 209 of the base station 201 is a multi-antenna antenna array comprising two or more antenna elements thereby allowing adaptive beam forms to be generated and directed in different directions based on the processing of the base station 201.

The signal received from the remote station 203 is filtered, amplified and down-converted to base-band as will be well known to the person skilled in the art (for clarity and brevity this functionality is not explicitly illustrated in FIG. 2) . The base-band signal is fed to a base station quantizer 211 which generates quantized base-band samples for the received signal. The quantized samples may be time-domain samples or frequency domain samples, obtained from the Fourier Transform of time- domain samples. The frequency-domain representation of samples is for example often used in Orthogonal Frequency Division Multiplexing (OFDM) systems. The quantized baseband samples are fed to a base station sample memory 213 wherein the samples are stored until the following time slot . CML04885M

The relay station 205 also comprises an antenna array 215 which receives the signal from the remote station 203. As for the base station 201, the antenna array 215 is a multi-antenna antenna array comprising two or more antenna elements thereby allowing adaptive beam forms to be generated and directed in different directions.

The signal received from the remote station 203 is filtered, amplified and down-converted to base-band as will be well known to the person skilled in the art (for clarity and brevity this functionality is not explicitly illustrated in FIG. 2) . The base-band signal is fed to a relay quantizer 217 which generates quantized base-band samples for the received signal. The quantized base-band samples are fed to a compressor 219 or which performs a lossy compression of the quantized base-band samples. The lossy compression is aimed at providing a more efficient representation of the quantized base-band samples without introducing significant errors. Thus, the lossy compression is a source encoding of the quantized baseband samples and is in the specific example a Wyner-Ziv source encoding. It will be appreciated, that in other embodiments other compression/ source encoding techniques may be used and specifically that a non-lossy compression may be used.

The relay compressor 219 is coupled to a relay sample memory 221 wherein the quantized and compressed base-band samples are stored until the following time slot.

The relay station 205 furthermore comprises a packet transmitter 223 coupled to the relay sample memory 221. CML04885M

The packet transmitter 223 is capable of transmitting a data packet to the base station 201. The base station comprises a packet receiver 221 which is capable of receiving this data packet and of receiving the signal from the remote station 203 based on this data packet and the base-band samples stored in the base station sample memory 213.

FIG. 3 illustrates the elements of FIG. 2 in the following time slot (henceforth referred to as the relay time slot) . In this time slot, the antenna array 215 of the relay station 205 is coupled to the packet transmitter 223 which transmits a data packet comprising the quantized and compressed base-band samples received from the remote station 203 in the previous time slot. The data packet is transmitted to the base station 201 where it is received by the antenna array 209.

In the relay time slot, the antenna array 209 is coupled to the packet receiver 225 which is furthermore coupled to the base-band sample memory 213. Thus, in the relay time slot the packet receiver 225 receives the base-band samples for all the antenna elements of the antenna array 215 of the relay station 205 as well as the base-band samples of all the antenna elements of the antenna array 209 of the base station 201.

The packet receiver 225 then proceeds to jointly process the base-band signals from the combined set of antenna elements made up by both the antenna elements of the base station antenna array 209 and the antenna elements of the relay station antenna array 215. It will be appreciated that various algorithms and techniques for jointly ° CML04885M processing base-band samples from a plurality of antenna elements to regenerate a transmitted signal will be well known to the person skilled in the art.

In the described cellular communication system, spatial separation is used to separate different remote stations using the same frequency-time (-code) resource block (henceforth referred to as a resource chunk) . Thus, the packet receiver 225 receives transmissions from a plurality of remote stations within each time slot.

Specifically, it generates a plurality of spatial data streams wherein each data stream corresponds to a beam aimed in specific direction of the remote station that has transmitted in the data.

Accordingly, the packet receiver 225 is operable to perform spatial separation joint processing of the baseband samples from the combined set of antenna elements. Specifically, the packet receiver 225 comprises functionality for implementing beamforming algorithms that can generate adaptable beams from the antenna arrays 209, 215. For example, the packet receiver 225 can adjust a gain and phase or delay for each antenna element for each spatial data stream in order to generate a desired beam direction for that spatial data stream.

It will be appreciated that various beamforming algorithms will be known to the person skilled in the art. For example, the packet receiver 225 may include a linear processing beamforming algorithm, such as the Minimum Mean Square Error algorithm, or a non-linear algorithm, such as the Successive Interference Cancellation algorithm. 14 CML04885M

Although some beamforming algorithms may be based on an assumption that the received base-band signals comprise only the wanted signals and additive white Gaussian noise which is completely uncorrelated between the different antenna elements, such an assumption is typically not valid in a practical system.

It has been found that improved performance can be achieved by taking specific information about the level and correlation of noise at the antenna elements into account. In the present application, the term noise will be used as a common term to include contributions from both thermal noise, interference, processing noise etc. Thus, specifically, the term noise includes both correlated and uncorrelated noise. Typically, the thermal noise will be uncorrelated between antenna elements whereas the interference may have a relatively high correlation .

In the described cellular communication system, the spatial processing is arranged to take into account the specific noise (and interference) conditions and characteristics. Furthermore, the spatial processing is arranged to not only take into account the conditions that are experienced at the base station 201 but also the conditions experienced at the relay station antenna array 215. Furthermore, the inventor has realised that the quantization and/or compression performed in the relay station 205 in order to reduce the data rate of the communication of the base-band samples to the base station 201 has a substantial impact on the noise conditions for the resulting base-band signals and ° CML04885M therefor has substantial impact on the spatial processing.

Accordingly, the spatial processing performed in the packet receiver 225 not only takes into account the noise conditions of the antenna arrays 209, 215 but also takes into account the effect of the data rate reduction processing (i.e. the quantization and compression) performed in the relay station 205. Thus, in the example the spatial processing performed by the packet receiver 225 takes into account at least one data rate parameter which is indicative of the data rate reduction processing performed at the relay station 205.

FIG. 4 illustrates an example of some elements of the packet receiver 225. FIG. 4 specifically illustrates a spatial processor 401 which performs the spatial separation joint processing of the base-band samples from the combined set of antenna elements 209, 215.

The spatial processor 401 specifically performs a beamforming algorithm to generate a plurality of spatial data streams (d) . The input to the spatial processor 401 includes the base-band samples from each antenna element. As illustrated in FIG. 4, the spatial processing is not necessarily limited to two antenna arrays but may be extended to any number of antenna arrays or relay stations. Thus, FIG. 4 shows an input 403 comprising the base-band samples for all antenna elements of the local antenna array 209, a separate input comprising the baseband samples for all antenna elements of the relay antenna array 215 (received in a separate transmission from the relay station 205) , as well as further optional " CML04885M inputs 407 for antenna arrays of other relay stations. For clarity and brevity, the following description will focus on examples wherein signals are received from only to antenna arrays .

The packet receiver 225 also comprises a parameter processor 409 which is coupled to the spatial processor 401. The parameter processor 409 generates a measure of the noise levels and correlations between the base-band samples for the different antenna elements. As mentioned previously, this measure does not only include contributions from the conditions at the antenna arrays 209, 215 but also includes a contribution from the data rate reduction processing of the relay station 205. Therefore, a data rate parameter indicative of this processing is provided to the parameter processor 409 as an input.

In the example, the spatial processor 401 receives a noise covariance matrix from the parameter processor 409. The number of columns and rows in the matrix corresponds to the total number of antenna elements in the combined set (i.e. the sum of the antenna elements in the first antenna array 209 and the second antenna array 215) .

The matrix coefficients of the diagonal of the noise covariance matrix represent the estimated noise level for the corresponding antenna element. The matrix coefficients outside the diagonal represent the cross- correlation between the noise of the two antenna elements corresponding to the matrix position. CML04885M

It will be appreciated that for the first antenna array 209 the noise and correlation introduced by the processing of the signals from the remote stations will typically be insignificant compared to the noise received by the antenna elements. For example, the quantisation of the base station quantizer 211 is typically sufficiently high to introduce only negligible quantisation noise which is furthermore uncorrelated between samples and antenna elements.

For the relay antenna array 215 the received thermal noise is also typically uncorrelated between antenna elements whereas the interference may have significant correlation. Furthermore, in contrast to the processing of the base station 201, the processing performed by the relay station 203 in order to reduce the data rate of the base-band samples forwarded to the base station 201 will typically have a significant impact on the characteristics of the resulting base-band samples. Specifically, the quantization and/or compression may be quite severe resulting a significantly increased noise and/or correlation. This additional correlated noise is in general different from the quadratic distortion introduced by the quantization and/or compression process.

Accordingly, the matrix coefficients of the noise covariance matrix corresponding to the antenna elements of the relay antenna array 215 takes into account at least one data rate parameter indicative of this data reduction process. 1 ft

° CML04885M

In the example, the data rate reduction processing of the relay station 205 includes the quantization performed by the relay station quantizer 217. Accordingly, the data rate parameter can comprise a quantization parameter indicative of the quantization performed by the relay station quantizer 217. For example, the data rate parameter can represent a quantization step used by the quantizer 217, a number of bits allocated to each baseband sample by the quantizer 217 and/or a type of quantization used by the quantizer 217 (such as whether a uniform or logarithmic quantization is used) .

Also, the data rate reduction processing of the relay station 205 includes the compression performed by the relay station compressor 219. Accordingly, the data rate parameter can comprise a compression parameter indicative of the compression performance of the relay station compressor 219.

For example, the operation of the packet receiver 225 can be dependent on a data rate parameter indicative of the type of compression which is used by the relay station compressor 219. For example, the generated noise covariance matrix may depend on whether a lossy or non lossy source encoding is used or can be dependent on which lossy source encoding is used. Thus, the noise covariance matrix may specifically reflect which source encoding is used and thus reflect the noise and correlation impact of using the specific source encoding.

Alternatively or additionally, the degree of compression may be taken into account. Thus, the compression rate (or degree) of the compression performed by the relay station -I Q σ CML04885M compressor 219 can be used to adjust the noise covariance matrix. The compression rate can typically impact the degree of lossyness in the compression and the higher the compression rate the higher is the introduced noise and/or correlation. This may be considered in the spatial separation joint processing performed by the packet receiver 225 thereby providing improved performance.

In the following, a more detailed description of how a combined noise covariance matrix for the first and second antenna arrays 209, 215 can be determined is described with reference to FIG. 5 which illustrates elements of the parameter processor 409 of FIG. 4.

The parameter processor 409 comprises a local noise covariance processor 501 which is operable to determine a noise covariance matrix for the first antenna array 209, i.e. for the local antenna array. The generated noise covariance matrix has a number of rows and columns that correspond to the number of antenna elements in the first antenna array 209.

The derivation of the noise covariance matrix for the local antenna array 209 may be in accordance with known principles.

For example, in a simple implementation the noise covariance matrix can be assumed diagonal and the diagonal terms may be obtained by measuring the noise variance for each antenna. In a more complex implementation, the system may also be subject to correlated interference, and the covariance matrix of the interference may be computed from the estimate of the ^ CML04885M multipath channel between the receiver and the interferer. The coefficients of this multipath channel may be obtained by channel sounding techniques, the latter relying for instance on the transmission of known pilot symbols by the interferer.

As an example, if H denotes the matrix of coefficients of the Multiple Input Multiple Output (MIMO) channel between the interferer and the receiving station on a given OFDM subcarrier, and if the interferer transmits at power P, then the interference covariance matrix at the receiver can be estimated as PHH^H where the ()^H operator denotes the Hermitian transpose. If the thermal noise covariance at the receiver is the diagonal matrix O²I where I ^ . ; x

O²I+ PHH^H.

The parameter processor 409 furthermore comprises a relay noise covariance processor 503 which is arranged to determine a first relay noise covariance matrix for the second antenna array 215, i.e. for the relay station antenna array. The first noise covariance matrix thus has a number of rows and columns that correspond to the number of antenna elements in the second antenna array 215.

The first noise covariance matrix represents the noise and correlation of the signals received by the individual antenna elements of the second antenna array 215. Thus, the first noise covariance matrix corresponds directly to the local noise covariance matrix determined by the local noise covariance processor 501 and can be determined CML04885M based on the measured noise and interference values for the individual antenna elements of the second antenna array 215 using the same principles as described for the local noise covariance matrix.

In addition the parameter processor 409 comprises a relay processing covariance processor 505 which determines a second noise covariance matrix for the second antenna array 215, i.e. for the relay station antenna array. The second relay noise covariance matrix therefore also has a number of rows and columns that correspond to the number of antenna elements in the second antenna array 215.

The second relay noise covariance matrix reflects the impact of the processing performed to reduce the data rate, i.e. the quantisation and compression of the relay base-band samples. Thus, based on the data rate parameter, the relay processing covariance processor 505 generates a second noise covariance matrix that reflects the noise and correlation contribution to the antenna elements of the second antenna array 215 introduced by the data rate reduction processing of the relay station 205.

It will be appreciated that the data rate parameter typically will be a combined parameter that includes a number of characteristics of the performed processing such as a quantization step, compression type, compression rate etc.

Thus, based on these parameters the relay processing covariance processor 505 determines the second noise covariance matrix for the second antenna array 215. The 22 CML04885M second noise covariance matrix may specifically be determined by first determining whether the compression process includes a linear transform, and if so, the specific transform must have been signaled. This linear unitary transform may for instance be the identity matrix, or a Karhunen-Loeve transform in a more complex implementation, or even a Conditional Karhunen-Loeve Transform in an even more complex implementation which corresponds to Wyner-Ziv compression when the relay has Multiple Antennas. This transform may be denoted by a matrix U.

The next step consists in determining at which rate each sample stream at the output of this linear transform was compressed. Based on the variance S₁ of each sample stream at the output of the linear transform and on the rate r_x at which this stream was compressed, a compression noise variance n_x can be computed. The relationship between S₁, -T₁ and T]₁ depends on the performance of the compression algorithm and may be pre-computed and stored in a look-up table. The additional noise covariance on the decompressed signal due to compression can be computed as the following matrix product U^Hdiag(η₁)U.

The first noise covariance matrix and the second noise covariance matrix for the relay antenna array 215 are fed to a matrix summation unit 507 which adds the two matrices together. The matrix summation unit 507 may simply add the individual matrix coefficients of the two matrices together. Thus, at the output of the matrix summation unit 507 a single noise covariance matrix for the relay antenna array 215 is generated with a number of ° CML04885M rows and columns that correspond to the number of antenna elements in the second antenna array 215.

The parameter processor 409 furthermore comprises a matrix processor 509 which is coupled to the matrix summation unit 507 and the local noise covariance processor 501. The matrix processor 509 receives the relay noise covariance matrix and the local noise covariance matrix and combines these to generate a combined noise covariance matrix.

The combined noise covariance matrix is a noise covariance matrix for the combined set of antenna elements of the first and second antenna arrays 209, 215 and thus has a number of rows and columns corresponding to the total number of antenna elements in the two antenna arrays 209, 215.

The combined noise covariance matrix is specifically generated by setting the matrix coefficients for antenna element pairs to the matrix coefficients of the relay noise covariance matrix and the local noise covariance matrix for the corresponding antenna element pairs. Thus, for two antenna elements belonging to the second antenna array 215, the matrix coefficient of the combined noise covariance matrix is set to the matrix coefficient for the same two antenna elements in the relay noise covariance matrix. Similarly, for two antenna elements belonging to the first antenna array 209 the matrix coefficient of the combined noise covariance matrix is set to the matrix coefficient for the same two antenna elements in the local noise covariance matrix. CML04885M

For the antenna element pairs which do not have a direct correspondence in either the relay noise covariance matrix or the local noise covariance matrix (i.e. for an antenna element pair comprising one antenna element from the first antenna array 209 and one antenna from the second antenna array 215) , the matrix coefficient is set zero reflecting that the noise correlation between the signals from the two antenna arrays 209, 215 is assumed to be uncorrelated.

The combined noise covariance matrix is then fed to the spatial processor 401 where it is used for the spatial separation joint processing, and specifically is used to generate beams separating different remote stations.

In some embodiments, an explicit indication of the data rate parameter indicative of the data rate reduction processing performed at the relay station 205 may be transmitted from the relay station 205 to the base station 201. For example, the data packet comprising the quantized and compressed base-band samples may include an indication of the type and degree of quantization used as well as the type and degree of compression used.

It will be appreciated that in other embodiments, the base station 201 may derive the appropriate parameters from the received data packet. For example, different types of source decoding may be applied to the received data packet with the appropriate source encoding/decoding algorithm being selected as the one resulting in the least errors. ° CML04885M

In some embodiments, the base station 201 may have prior knowledge of the data rate reduction used by the relay station, for example it may be known which compression algorithm and/or degree is used and the packet receiver 225 and the base station may automatically take this into account .

It will be appreciated that in many embodiments, the data rate parameter set taken into account when performing the spatial separation joint processing is exactly the same data rate parameter set required to decode the received relay station base-band samples. E.g. knowledge of the compression type and degree is required in order to decompress the received data packet.

It will also be appreciated that the described processing can be implemented in different locations. For example, the relay noise covariance matrix may be calculated at the base station 201 as described previously. However, as another example, the relay noise covariance matrix may be determined in the relay station 205 and transmitted to the base station 201 e.g. by including it in the data packet carrying the relay station base-band samples.

In the previously described example, a base station 201 cooperated with a single relay station 205. However it will be appreciated that in other embodiments more than one relay station may be involved (as e.g. is indicated by the additional blocks of FIG. 4 and FIG. 5) .

It will also be appreciated that in some embodiments the relay station 205 may itself be a base station of the cellular communication system. In this case, the ° CML04885M transmission of the relay station base-band samples from the relay station 205 to the base station 201 may for example be via the fixed network of the cellular communication system (i.e. a backhaul network may be used instead of a direct air interface communication) .

As another example, the apparatus generating the spatial data streams may be separate from the actual relay stations and/or base stations. For example, the relay stations and/or base stations cooperating to receive the signals from the remote stations may all transmit the received base-band data to a spatial processor e.g. located in the fixed network of the cellular communication system. In such cases, all involved receivers may include data reduction processing and accordingly a noise covariance matrix taking into account this data reduction processing may be generated for each receiver with the combined noise covariance matrix then being generated by combining these individual noise covariance matrices.

In some embodiments, the communication system may also comprise a scheduler which is arranged to schedule which remote stations are allowed to use each resource chunk. Thus, in the specific example such a scheduler will be an SDMA scheduler that allows a plurality of remote stations to use the same resource chunk provided they can be spatially separated by the beams generated by the spatial separation joint processing.

Accordingly, the scheduler takes into account the data rate parameter indicative of the data rate reduction CML04885M performed at the relay station 205 since this processing affects the spatial separation that can be achieved.

In the specific example, the scheduler schedules remote stations for a given resource chunk in response to the spatial separation joint processing performed in the base station 201. The scheduler evaluates the spatial separation joint processing for different potential scheduling options (e.g. for a scheduling of different sets of remote stations) . The resulting output of the spatial processor for each of these sets is then evaluated and the set resulting in the highest combined data rate is selected.

Thus, the approach corresponds to a sum-rate scheduler (which seeks to maximises the sum of the data rates that are scheduled in any given resource chunk) . The sum rate maximisation is based on the output of the spatial processor which depends on the data rate parameter and accordingly the scheduling is dependent on both the spatial separation joint processing and the data rate parameter reflecting the data reduction processing performed at the relay station 205.

Thus, the described approach allows an improved system wherein SDMA is combined with relaying of signals. Significantly improved resource efficiency and capacity of the cellular communication system can be achieved as well as an improvement in the quality of service for each individual communication service. Furthermore, the described system allows relaying based on Compress-and- Forward to be combined with SDMA thereby resulting in the benefits and advantages of both systems being combined. ° CML04885M

In particular, the described approach allows a substantially higher exploitation of spatial separation in order to increase the capacity of the system. Specifically, the number of individual users that can be spatially separated within each resource chunk is restricted by the sum of antenna elements in all the involved antenna arrays which rather than being limited by the number of antenna elements in a single antenna array.

FIG. 6 illustrates a method of receiving data from remote stations .

The method initiates in step 601 wherein first signals received by a first antenna array from a plurality of remote stations are received.

Step 601 is followed by step 603 wherein second signals received by a second antenna array of a first relay station from the plurality of remote stations is received. The second signals are represented by data dependent on a data rate parameter indicative of a data rate reduction processing performed at the first relay station.

Step 603 is followed by step 605 wherein a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals is performed to generate received data streams for a set of remote stations of the plurality of remote stations. The spatial separation joint processing provides a spatial separation for the received data streams from the set of 2Q

^Ό CML04885M remote stations and the spatial separation joint processing is in response to the data rate parameter.

It will be appreciated that the above description for clarity has described embodiments of the invention with reference to different functional units and processors. However, it will be apparent that any suitable distribution of functionality between different functional units or processors may be used without detracting from the invention. For example, functionality illustrated to be performed by separate processors or controllers may be performed by the same processor or controllers. Hence, references to specific functional units are only to be seen as references to suitable means for providing the described functionality rather than indicative of a strict logical or physical structure or organization.

The invention can be implemented in any suitable form including hardware, software, firmware or any combination of these. The invention may optionally be implemented at least partly as computer software running on one or more data processors and/or digital signal processors. The elements and components of an embodiment of the invention may be physically, functionally and logically implemented in any suitable way. Indeed the functionality may be implemented in a single unit, in a plurality of units or as part of other functional units. As such, the invention may be implemented in a single unit or may be physically and functionally distributed between different units and processors . ^oυ CML04885M

Although the present invention has been described in connection with some embodiments, it is not intended to be limited to the specific form set forth herein. Rather, the scope of the present invention is limited only by the accompanying claims. Additionally, although a feature may appear to be described in connection with particular embodiments, one skilled in the art would recognize that various features of the described embodiments may be combined in accordance with the invention. In the claims, the term comprising does not exclude the presence of other elements or steps.

Furthermore, although individually listed, a plurality of means, elements or method steps may be implemented by e.g. a single unit or processor. Additionally, although individual features may be included in different claims, these may possibly be advantageously combined, and the inclusion in different claims does not imply that a combination of features is not feasible and/or advantageous. Also the inclusion of a feature in one category of claims does not imply a limitation to this category but rather indicates that the feature is equally applicable to other claim categories as appropriate. Furthermore, the order of features in the claims does not imply any specific order in which the features must be worked and in particular the order of individual steps in a method claim does not imply that the steps must be performed in this order. Rather, the steps may be performed in any suitable order.

Claims

° CML04885MCLAIMS

1. An apparatus comprising: means for receiving first signals received by a first antenna array from a plurality of remote stations; means for receiving, from a first relay station, second signals received by a second antenna array of the first relay station from the plurality of remote stations, the second signals being represented by data dependent on a data rate parameter indicative of a data rate reduction processing performed at the first relay station; means for performing a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals to generate received data streams for a set of remote stations of the plurality of remote stations, the spatial separation joint processing providing a spatial separation for the received data streams; and wherein the spatial separation joint processing is in response to the data rate parameter.

2. The apparatus of claim 1 wherein the data rate parameter comprises a quantization parameter for a quantization performed by the first relay station.

3. The apparatus of claim 1 wherein the data rate parameter comprises a data compression parameter for a data compression performed by the first relay station. ° CML04885M

4. The apparatus of claim 3 wherein the data compression parameter comprises an indication of a type of data compression being performed by the relay station.

5. The apparatus of claim 3 wherein the data compression is a lossy compression.

6. The apparatus of claim 5 wherein the data compression parameter comprises an indication of a compression rate for the data compression.

7. The apparatus of claim 1 further comprising means for receiving an indication of the data rate parameter from the relay station.

8. The apparatus of claim 1 further comprising matrix means for generating a noise covariance matrix for the combined set of signals, the noise covariance matrix being dependent on the data rate parameter, and wherein the spatial separation joint processing is in response to the noise covariance matrix.

9. The apparatus of claim 1 wherein the matrix means is arranged to generate the noise covariance matrix by combining a first noise covariance matrix for the first signals and a second noise covariance matrix for the second signals, the second noise covariance matrix being dependent on the data rate parameter.

10. A method of receiving data from remote stations, the method comprising: receiving first signals received by a first antenna array from a plurality of remote stations; °° CML04885M receiving, from a first relay station, second signals received by a second antenna array of the first relay station from the plurality of remote stations, the second signals being represented by data dependent on a data rate parameter indicative of a data rate reduction processing performed at the first relay station; performing a spatial separation joint processing of a combined set of signals comprising the first signals and the second signals to generate received data streams for a set of remote stations of the plurality of remote stations, the spatial separation joint processing providing a spatial separation for the received data streams; and wherein the spatial separation joint processing is in response to the data rate parameter.